Monte-Carlo-simulation-based optimization for copolymerization processes with embedded chemical composition distribution

As chemical composition distribution (CCD) is a crucial microstructural quality index of copolymers, optimization of operating policies using CCD is of great importance. Monte Carlo simulation is an efficient method to calculate the CCD that cannot be easily determined by traditional equation-based methods But this method is computationally expensive. In this project, we first propose a parallel technique to conduct the Monte Carlo simulation on the graphics processing unit (GPU) platform. Additionally, an adaptive simulation algorithm is proposed to reduce computational cost based on error estimation of the Monte Carlo simulation. Considering the uncertainties in the Monte Carlo simulation, derivative-free method is applied for the CCD-target optimization. A successive boundary shrinkage (SBS) formulation is developed to improve the convergence of problem solving. The above-mentioned methods are successfully integrated and implemented on the optimization of a copolymerization process with high efficiency and good performance.